carbon-rich chondritic clast pv1 from the plainview ... - uji · carbon-rich chondritic clast pv1...

Geochimica et Cosmochimica Acta, Vol. 69, No. 13, pp. 3419–3430, 2005Copyright © 2005 Elsevier Ltd

Printed in the USA. All rights reserved

doi:10.1016/j.gca.2004.11.001

Carbon-rich chondritic clast PV1 from the Plainview H-chondrite regolith breccia:Formation from H3 chondrite material by possible cometary impact

ALAN E. RUBIN,1,* JOSEP M. TRIGO-RODRIGUEZ,1 TAKUYA KUNIHIRO,1 GREGORY W. KALLEMEYN,1 and JOHN T. WASSON1,2,3

1Institute of Geophysics and Planetary Physics,2Department of Earth and Space Sciences, and

3Department of Chemistry and Biochemistry, University of California, Los Angeles, CA 90095-1567, USA

(Received April 29, 2004; accepted in revised form November 2, 2004)

Abstract—Chondritic clast PV1 from the Plainview H-chondrite regolith breccia is a subrounded, 5-mm-diameter unequilibrated chondritic fragment that contains 13 wt% C occurring mainly within irregularlyshaped 30–400-�m-size opaque patches. The clast formed from H3 chondrite material as indicated by themean apparent chondrule diameter (310 �m vs. �300 �m in H3 chondrites), the mean Mg-normalizedrefractory lithophile abundance ratio (1.00 � 0.09�H), the previously determined O-isotopic composition(�17O � 0.66‰ vs. 0.68 � 0.04‰ in H3 chondrites and 0.73 � 0.09‰ in H4-6 chondrites), the heteroge-neous olivine compositions in grain cores (with a minimum range of Fa1-19), and the presence of glass insome chondrules. Although the clast lacks the fine-grained, ferroan silicate matrix material present in type 3ordinary chondrites, PV1 contains objects that appear to be recrystallized clumps of matrix material. Similarly,the apparent dearth of radial pyroxene and cryptocrystalline chondrules in PV1 is accounted for by thepresence of some recrystallized fragments of these chondrule textural types. All of the chondrules in PV1 areinterfused indicating that temperatures must have briefly reached �1100°C (the approximate solidus temper-ature of H-chondrite silicate). The most likely source of this heating was by an impact. Some metal was lostduring impact heating as indicated by the moderately low abundance of metallic Fe-Ni in PV1 (�14 wt%)compared to that in mean H chondrites (�18 wt%). The carbon enrichment of the clast may have resulted froma second impact event, one involving a cometary projectile, possibly a Jupiter-family comet. As the clastcooled, it experienced hydrothermal alteration at low water/rock ratios as evidenced by the thick rims offerroan olivine around low-FeO olivine cores. The C-rich chondritic clast was later incorporated into the

0016-7037/05 $30.00 � .00

H-chondrite parent-body regolith and extensively fractured and faulted. Copyright © 2005 Elsevier Ltd

1. INTRODUCTION

Carbon is a minor component of chondritic meteorites: type3 ordinary chondrite (OC) falls typically contain 0.3–0.6 wt%C, two CV3 falls (Allende and Bali) also contain 0.3–0.6 wt%C, CM2 falls contain �1.8 wt% C, and CI Orgueil contains 2.8wt% C (Jarosewich, 1990). Poorly graphitized carbon occurswithin metallic Fe-Ni grains and as relatively abundant sub-millimeter-size carbon-rich aggregates in some type 3 OC(McKinley et al., 1981; Scott et al., 1981a; Brearley, 1990;Mostefaoui et al., 2000). Far more C (6–13 wt%) is present infour enigmatic chondritic clasts described by Scott et al.(1981a,b). The clasts range in diameter from 1 to 11 mm andoccur in H-chondrite regolith breccias: PV1 from Plainview(1917) (hereafter Plainview), DT1 and DT2 from Dimmitt, andWN1 from Weston. The clasts have also been reported to lack thefine-grained ferroan-silicate-rich matrix material (Scott et al.,1981a,b, 1988; Brearley, 1990) that occurs ubiquitously in type 3ordinary (Huss et al., 1981) and carbonaceous (McSween andRichardson, 1977) chondrites.

Previous studies have concluded that the C-rich chondriticclasts formed in the solar nebula and are thus most likelycomposed of primitive material (Scott et al., 1981b, 1988;Brearley, 1990). In contrast, the present study of the PV1chondritic clast indicates that most features were produced byasteroidal processes.

* Author to whom correspondence should be addressed([email protected]).

3419

2. ANALYTICAL PROCEDURES

A significant fraction of clast PV1 occurs within University-of-New-Mexico thin section UNM 273 of the Plainview H-chondrite regolithbreccia (Fig. 1a). We studied this thin section microscopically in transmit-ted and reflected light and prepared a mosaic back-scattered electron (BSE)image of the clast. A grid was superimposed on this image; labels ofchondrules and other objects in the image reflect their location on this grid.All BSE images were made with the LEO 1430 VP scanning electronmicroscope (SEM) at UCLA using a 15 keV acceleration voltage and aworking distance of �19 mm. Chondrule and grain sizes were measuredon the BSE images. Mineral compositions were determined with the JEOLJXA-8200 electron microprobe at UCLA using natural and syntheticstandards, an accelerating voltage of 15 keV, a 15-nA sample current, 20-scounting times, and ZAF corrections.

A 5.46-mg chip of PV1, obtained from the University of New Mexico,was separated for instrumental neutron activation analysis (INAA). Thesample was irradiated at the University of California, Irvine (UCI) reactorwith a neutron flux of �1.8 � 1012 cm�2 s�1. Samples were irradiated fortwo minutes and then counted immediately to determine elements produc-ing very short-lived radioisotopes (Mg, Al, Ca, V and Mn). Samples wereirradiated again for 4 h and counted several times over a period of �6weeks to determine elements producing longer-lived species (Na, K, Ca,Sc, Cr, Mn, Fe, Co, Ni, Zn, Ga, As, Se, La, Sm, Eu, Tb, Ho, Yb, Lu, Ir andAu). The INAA procedure is described in Kallemeyn et al. (1989). Theestimated relative sample standard deviations are 6%–10% for K, Eu, Yband Lu, and �5% for all other elements.

3. RESULTS

3.1. Petrography and Mineralogy

3.1.1. Plainview whole rock

Plainview is an H-chondrite regolith breccia (fig. 1 of Fodor and
Keil, 1976) containing solar-wind-implanted rare gases (Schultz

3420 A. E. Rubin et al.

and Kruse, 1989). The meteorite was found in Hale County,Texas; �700 kg have been recovered (Grady, 2000). Plainviewcontains �30 vol% H5 clasts, �1 vol% light-colored impact-melt-rock clasts, �1 vol% exotic clasts (mainly phyllosilicate-bearing CM2 chondrite fragments), and �1 vol% shocked (butunmelted) H-chondrite clasts of different petrologic types (Fodorand Keil, 1976; Rubin, 1982); the remaining �70 vol% consists of

Fig. 1. Clast PV1. (a) Back-scattered electron (BSE) imaused in locating and identifying components is superposedsection is shown at lower left. (b) Portion of a mosaic of thabundant C-rich aggregates (black) and fused chondruleSmall scale bar at upper left is 100 �m long. (c) Reflected(white) penetrating the clast from the Plainview host.

clastic chondritic material surrounding recognizable clasts.

The Plainview whole rock comprises subrounded �0.02–40-mm-size clasts surrounded by glassy-to-microcrystallinefeldspathic material (Bischoff et al., 1983). Plainview wasclassified as shock-stage S3, indicating that the examined por-tions of the breccia experienced a peak shock pressure of 5–10GPa (Stöffler et al., 1991). Plainview contains opaque shockveins, melt pockets and weakly developed melt dikes (Stöffler

e PV1 clast and the surrounding Plainview host. The gridimage. The location of the clast relative to the entire thin

mage of PV1 within the Plainview host. The clast containss of gray) of different textural types and compositions.mage of PV1 showing prominent veins of metallic Fe-Ni

ge of thon the

e BSE is (shade

light i

et al., 1991).

b

3421Carbon-rich chondritic clast from Plainview

The numerous impact-melt-rock clasts in Plainview are gen-erally very similar in bulk composition to the silicate portion ofthe Plainview host, consistent with the view that they formedfrom the host by impact melting, separation of a dense, immis-cible metal-sulfide liquid from the silicate liquid, and rapidcooling of the residual silicate melt (Fodor and Keil, 1976). The39Ar-40Ar age of one 1.5 � 4-cm-size impact-melt-rock clast is3.63 Ga; the age of the Plainview host is 4.4 Ga (Keil et al.,1980).

3.1.2. Clast PV1

PV1 is a subrounded 5-mm-diameter dark-colored chondriticclast (Fig. 1a,b) that contains �13 wt% carbonaceous matter(predominantly poorly graphitized C) occurring mainly withinirregularly shaped 30–400-�m-size opaque patches (Fodor andKeil, 1976; Scott et al., 1981a,b, 1988; Brearley, 1990; thisstudy). Modal analysis indicates that the clast contains �6vol% (i.e., �14 wt%) metallic Fe-Ni (Scott et al., 1981b);however, this is an upper limit because some of the metaloccurs in veins (particularly two prominent subparallel 2–2.5-mm-long veins; Fig. 1c) that are not indigenous to the clast(Scott et al., 1981b; this study).

3.1.2.1. Mineral compositions. Olivine and low-Ca py-roxene are compositionally heterogeneous: mean Fa � 17 � 8mol% (PMD � 46); mean Fs � 10.6 � 6.3 mol% (PMD � 48)(Scott et al., 1981b). These compositional distributions (Fig. 2)roughly resemble those of moderately primitive (type �3.6)OC (e.g., fig. 1 of Dodd et al., 1967). The paucity of Fa 0–2grains in the PV1 olivine distribution is most similar to that ofH/L3.6 Tieschitz (fig. 1 of Dodd et al., 1967).

3.1.2.2. Chondrules. Chondrules, chondrule fragmentsand coarse silicate grains together constitute 63 vol% of PV1(Scott et al., 1981b). Most chondrules are type I and type IIporphyritic olivine (PO) and porphyritic olivine-pyroxene(POP). Two barred olivine (BO) chondrules (e.g., Fig. 3a) anda few small granular olivine-pyroxene (GOP) chondrules occurin the thin section we studied. Very rare enveloping compoundchondrules and radial pyroxene (RP) chondrules are alsopresent. No readily recognizable cryptocrystalline chondruleswere encountered. Glassy mesostasis occurs between the barsin the two BO chondrules (e.g., Fig. 3a) as well as betweenphenocrysts in some PO and POP chondrules. We measured theapparent diameters of 60 chondrules in PV1 and found a meanapparent diameter of 310 � 120 �m.

3.1.2.3. Fractures. Numerous 1–3-�m-thick, 20–160-�m-long, cross-cutting fractures transect every chondrule in thePV1 clast (e.g., Fig. 3b). The fractures appear bright in BSEimages, reflecting the fact that they are filled with thin veins ofopaque material: some veins contain troilite, some containgoethite, and some contain both phases. (Goethite is a terres-trial weathering product.) In a few cases, troilite and goethiteoccur together as adjacent parallel veins filling the samefracture.

3.1.2.4. Chondrule fusion. Most interfaces between adja-
cent chondrules are melded together (Fig. 4). The boundaries
between many chondrules are identifiable only because of thechange in texture and grain size between adjacent chondrules ofdistinct textural types. Mineral grains at the interfaces betweendifferent chondrules abut each other, in most cases with nodiscernable gap. Although some chondrules are partly sur-rounded by C-rich patches, other parts of the perimeters ofthese chondrules are finely intergrown with those of adjacentchondrules. The chondrule boundaries resemble those of com-pound chondrules (figs. 1–3 of Wasson et al., 1995) that fusedin the nebula.

3.1.2.5. Recrystallized nonporphyritic chondrules. Al-though few RP chondrules and no cryptocrystalline chondruleswere identified in the available thin section of PV1, there areobjects that appear to be recrystallized fragments of thesetextural types. For example, chondrule E6 (Fig. 5a) is a dis-torted-oval-shaped 230 � 370-�m-size object that consists ofat least two distinct sets of subparallel irregular bars of low-Capyroxene (Fs16.7Wo3.3); the bars range in length from �70–120 �m and in thickness from 1.8–4 �m. Small (1–4-�m-

Olivine (n = 275)

0

5

10

15

20

25

1

Fa (mol%)

Nu

mb

er

of

Gra

ins

Low-Ca Pyroxene (n = 241)

0

5

10

15

20

25

30

35

1

Fs (mol%)

Nu

mb

er

of

Gra

ins

a

Fig. 2. Compositional distributions of mafic minerals in clast PV1determined by random electron microprobe analysis of �5-�m-sizegrains. (a) Olivine (n � 275 grains) is very heterogeneous, but there isa paucity of grains at very low Fa contents. The relative dearth of grainsat Fa19-20 is probably an artifact. (b) Low-Ca pyroxene (n � 241grains) is also very heterogeneous, but there is a paucity of grains atlow Fs contents. As in typical highly unequilibrated type 3 OC, themean Fs content has a lower numerical value than the mean Fa content.Data from Scott et al. (1981b).

thick) patches of interstitial glass occur between the bars. The


sets of bars are inclined with respect to each other at an angleof �40°. A few 4–20-�m-size anhedral and subhedral low-Capyroxene grains are present between some of the bars. E6 isinferred to be a recrystallized RP chondrule.

Chondrule D2 (Fig. 5b) appears to be a recrystallized cryp-tocrystalline chondrule fragment. It is 720 �m in diameter andconsists of 2–3-�m-size grains of low-Ca pyroxene(Fs38.8Wo1.4) and rare subcalcic augite (Fs23.1Wo23.1) sep-arated by �1-�m-size voids. In many places, the small pyrox-ene grains appear to be aligned to form �100-�m-long, 2–3-�m-thick pyroxene bars. In different regions the bars areoriented in different directions. Thin troilite veins (typically 1� 10 �m) occur between some of the bars. Numerous fracturesfilled with goethite transect the chondrule.

3.1.2.6. Recrystallized fine-grained silicate matrix material.

Fig. 3. Chondrules. (a) Rimmed barred olivine (BO) chondrule B4containing abundant glassy mesostasis between the olivine bars. Sev-eral bars and grains near the rim have low-FeO cores surrounded byhigh-FeO olivine. A few fractures (white) transect the chondrule. Scalebar at lower left is 20 �m in length. (b) Porphyritic olivine (PO)chondrule G4 transected by numerous fractures (white) filled withgoethite and/or troilite. Both images in BSE.

Although patches of fine-grained, FeO-rich, interchondrule sil-

icate matrix material are ubiquitous in type 3 OC (Huss et al.,1981), this material appears to be absent from PV1 (as well asthe other C-rich chondritic clasts) (Scott et al., 1981a,b). Thereare, however, a few objects in PV1 that appear to be recrystal-lized clumps of matrix material. One such object (I5; Fig. 5c)consists of �20 vol% small (1–17 �m), moderately angulartaenite and troilite grains, �30 vol% pore space presently filledwith C-rich material, and �50 vol% small (0.5–3 �m) silicategrains. The silicates are mainly ferroan olivine (Fa36.0) withminor subcalcic augite (Fs18.4Wo25.2). Rare grains of low-Capyroxene (Fs12.4Wo0.28) are also present. Several fracturestransect the object.

3.1.2.7. Ferroan olivine in chondrules. Many porphyriticchondrules, porphyritic chondrule fragments and isolated oli-vine grains in PV1 possess slightly fractured relict cores oflow-FeO olivine surrounded by moderately fractured ferroanolivine patches (Fig. 6). In many cases, ferroan olivine halossurround fractures in low-FeO olivine cores in the chondrules.

Fig. 4. Fused chondrules. (a) Region D4 consisting of intergrownporphyritic chondrules of different compositions and grain sizes. (b)Region I5 consisting of intergrown porphyritic chondrules interspersedwith small C-rich aggregates (black).

For example, object C4 (Fig. 6a) is a 175 � 215-�m-size POP

ost of t


chondrule fragment containing 6 � 9–75 � 115-�m-size fer-roan olivine grains (Fa32.7–35.7), 20–30-�m-size grains oflow-Ca pyroxene (Fs15.1Wo4.3) and pigeonite (Fs15.0Wo8.9),and interstitial mesostasis. The largest olivine grain consists ofseveral 5–18-�m-size relict relatively low-FeO (Fa12.0) coressurrounded by thick patches (up to 45 �m thick) of ferroanolivine (Fa35.7). The outer portion of the grain is composedentirely of ferroan olivine; 2–3-�m-thick bands of ferroanolivine also surround the troilite veins that fill the fractures.

Object D3 (Fig. 6b) is a wedge-shaped PO chondrule frag-ment, 220 � 360 �m in size. It contains coarse (40 � 60–105� 205 �m) olivine grains consisting of �20 vol% small (7 �11–25 � 35 �m) relatively low-FeO patches (Fa19.4) sur-rounded by ferroan olivine (�80 vol%; Fa32.5). The outerparts of the grains consist entirely of ferroan olivine.

Fig. 5. Recrystallized chondrules and fine-grained matcontaining at least two sets of distorted pyroxene bars andRecrystallized cryptocrystalline chondrule D2 consistingpresent filled with C-rich aggregate material). (c) Recrystanumerous grains of metallic Fe-Ni and sulfide (white); m

Object K3 is a 290 � 300-�m-size isolated olivine grain

consisting of relict relatively low-FeO (Fa13.5) zones rangingin size from 10 � 10 �m to 50 � 110 �m, hemmed in byfractures filled with troilite veins and surrounded by ferroanolivine (Fa30.1) zones up to 80 � 105 �m in size. Thin(2–10-�m-thick) patches of ferroan olivine surround most ofthe troilite veins.

Enveloping compound-chondrule G7 (Fig. 6c) consists of a200-�m-diameter recrystallized RP core (with Fs7.1Wo1.7low-Ca pyroxene grains) surrounded by a �140-�m-thickspherical shell consisting mainly of small (3–9-�m-size) eu-hedral olivine grains surrounded by mesostasis. Many of thesegrains have 3–5-�m-size low-FeO relict cores (Fa9.4–13.8)surrounded by 1–1.5-�m-thick ferroan olivine rims (Fa23.6–29.0) (Fig. 6d). Many of the euhedral olivine grains aretransected by 0.7-�m-wide fractures filled with ferroan olivine.

Partly recrystallized radial pyroxene (RP) chondrule E6itial glass. Numerous fractures transect the chondrule. (b)ll pyroxene grains separated by small voids (that are atlump of fine-grained silicate matrix material I5 containinghe silicate (dark gray) consists of ferroan olivine.

rix. (a)interstof sma

llized c

Barred olivine chondrule B4 (Fig. 3a) contains several oli-

re the lo


vine grains within the rim that consist of 3–6-�m-wide low-FeO cores (Fa1.2) surrounded by 1–2-�m-thick rinds of ferroanolivine (Fa21-25). Some of the olivine bars in the chondrulehave moderately low-FeO cores rimmed by ferroan olivine.Some small, relatively equant bars consist completely of fer-roan olivine; this is also the case for elongated olivine bars thathave a half-width of 1–2 �m.

3.1.2.8. Faults. Several faults with a maximum length of�1.2 mm and a maximum displacement of �200 �m occur inPV1 (Figs. 6c,d, 7). The faults are oriented in different direc-tions. One fault cuts across (and thus postdates) the fractures inchondrule F6, but another fault is transected by (and thuspredates) the fractures in compound chondrule G7 (Fig. 6c).The fault that cuts across G7 has a wedge-shaped cross-sectionand varies in thickness from 1.8 to 4.5 �m (Fig. 6d). One

Fig. 6. Ferroan olivine in chondrules. (a) PO chondrusurrounded by abundant ferroan olivine (light gray). FWedge-shaped PO chondrule fragment D3 containing larg(dark gray) surrounded by abundant ferroan olivine (lightRP core surrounded by a PO spherical shell containing maby thin rims of ferroan olivine. The chondrule is bisected b(d) Higher-magnification view of the fault transecting com�m and has a displacement of �180 �m. Also visible aeuhedral olivine grains.

�6-mm-long fault through the Plainview host transects the

entire PV1 clast, dividing it into two unequal segments (�20and �80 vol%) (fig. 1a of Scott et al., 1981b; fig. 10 of Rubin,1982). This particular fault is subparallel to two prominentmetal veins that penetrate the clast from the Plainview host.

3.1.2.9. Carbon-rich patches. The abundant carbon inPV1 occurs mainly within irregularly shaped 30–400-�m-sizeopaque patches (Fig. 1b). Previous TEM studies showed thepresence of poorly graphitized or turbostratic C associated withamorphous carbon (Brearley et al., 1987). Small (2–5 � 50–80nm) carbon crystallites are present, but well-crystallized graph-ite is not (Brearley et al., 1987; Brearley, 1990). Approximately10 wt% Fe is present within the C (Scott et al., 1981a,b), but theFe does not occur as a distinct crystalline phase (Brearley et al.,1987).

There is no evidence of silicate reduction at the interfaces

ment C4 consists of low-FeO olivine cores (dark gray)olivine also flanks most of the fractures (white). (b)crysts consisting of small residual low-FeO olivine coresc) Compound chondrule G7 consisting of a recrystallizedll euhedral olivine grains with low-FeO cores surroundedminent fault. Scale bar at lower right is 100 �m in length.chondrule G7. The fault varies in width from 1.8 to 4.5w-FeO cores and surrounding ferroan rims of the small

le fragerroane phenogray). (ny smay a propound

between the C-rich patches and adjacent silicate. Even small

2 conta


(3–10 �m) olivine grains completely embedded within the Cshow no such evidence. (If such reduction had occurred, por-tions of silicate grains adjacent to the C patches would showcompositional gradients similar to those observed adjacent tographite veins in ureilites, e.g., Berkley et al., 1976.) One mightalso expect “dusty olivine” (consisting of small blebs of Ni-poor metallic Fe) of the sort present in chondrules from type 3chondrites (e.g., fig. 5b of Jones, 1996) to occur in PV1, but nosuch structures are present.

In several places (e.g., F6, I4, K2), the BSE images showbright-shaded veins within the C-rich material that trace thefaults. In F6, a 1.8-�m-thick, 30-�m-long goethite vein withinan 80 � 440-�m-size C-rich patch is bent at an angle of �110°along the trajectory of a fault (Fig. 7a). In I4, a 1–2-�m-thick,170-�m-long curvilinear vein of troilite lies along the trace ofa fault within a 55 � 210-�m-size C-rich patch (Fig. 7b). InK2, a 110-�m-long, 1.4-�m-thick taenite vein flanked in placesby a 5-�m-thick mantle of goethite lies along the trace of a

Fig. 7. Faults in chondrules. (a) Coarse-grained PO chdisplacement on the fault is �50 �m. At the top center isgoethite vein bent at an angle of �110° along the trajectorvein of troilite along the trace of a fault. (c) Aggregate Kmantle of goethite along the trace of a fault.

fault within a 140 � 220-�m-size C-rich patch (Fig. 7c).

3.2. Bulk Chemical Composition

The concentrations of 25 elements in PV1, determined byINAA, are listed in Table 1. The Mg-normalized abundanceratios of these elements in PV1 relative to mean H-groupchondrites are shown in Figure 8.

Refractory lithophile elements are all within 17% of Hchondrites; they range from 0.86�H for Sm to 1.17�H for Ca.The mean refractory-lithophile-element abundance ratio for 10elements (Al, Sc, Ca, REE) is 1.00 � 0.09�H.

PV1 is depleted in the moderately volatile lithophile ele-ments Na (0.75�H) and K (0.70�H).

Iridium, the only refractory siderophile element that wasanalyzed, is at 1.2�H. Common siderophile elements are allappreciably depleted: Ni (0.70�H), Co (0.43�H), Fe(0.67�H). In contrast, the volatile siderophile elements arerelatively unfractionated: Au (1.05�H), As (0.93�H), Ga(1.10�H). Chalcophile Se (1.28�H) and Zn (1.38�H) abun-

fragment F6 transected by a fault oriented NW-SE. The440-�m-size C-rich aggregate containing a 30-�m-long

ault. (b) Aggregate I4 contains a 170-�m-long curvilinearins a 110-�m-long taenite vein flanked by a 5-�m-thick

ondrulea 80 �

y of a f

dances are moderately high in the PV1 clast.

Tab

le1.

Con

cent

ratio

nsof

25el

emen

tsde

term

ined

byIN

AA

ina

5.46

-mg

chip

ofPl

ainv

iew

clas

tPV

1a .

CN

aM

gA

lK

Ca

ScV

Cr

Mn

FeC

oN

iZ

nG

aA

sSe

La

SmE

uT

bH

oY

bL

uIr

Au

stPV

113

05.

6517

213

.066

517

.69.

2785

.54.

383.

1021

943

513

.976

7.7

2.41

12.6

350

200

9767

8026

238

1160

277

hond

rite

s1.

16.

2614

211

.478

212

.47.

8873

.33.

672.

3227

183

116

.345

.55.

782.

138.

130

319

174

5373

208

31.9

791

218

Con

cent

ratio

nsar

ein

�g/

gex

cept

:N

a,M

g,A

l,C

a,C

r,M

n,Fe

,Ni

inm

g/g;

RE

E,I

r,A

uin

ng/g

.Als

ogi

ven

isth

eC

conc

entr

atio

n(i

nm

g/g)

ofPV

1es

timat

edby

Scot

tet

al.(

1981

a,b)

.H-c

hond

rite

afr

omK

alle

mey

net

al.

(198

9)ex

cept

for

C,

Tb

and

Ho

(fro

mW

asso

nan

dK

alle

mey

n,19

88).


4. DISCUSSION

4.1. H3 Chondrite Precursor

4.1.1. H-group chondrite precursor

There are several characteristics of clast PV1 that indicatethat it formed from H-group chondrite material. Chondrules inPV1 have a mean apparent diameter of 310 � 120 �m, essen-tially identical to that of mean H3-chondrite chondrules (�300�m; Rubin, 2000). (For comparison, L and LL chondrites havemean chondrule diameters of �500 and �570 �m, respec-tively; Rubin, 2000; Nelson and Rubin, 2002.) The PV1 INAAdata show a mean Mg-normalized refractory lithophile abun-dance ratio (Al, Sc, Ca, La, Sm, Eu, Tb, Ho, Yb, Lu) relativeto H chondrites of 1.00 � 0.09, i.e., indistinguishable from thatof mean H chondrites. The O-isotopic composition of PV1(�18O � 3.35‰; �17O � 2.40‰; Scott et al., 1988) yields a�17O value of 0.66‰, within one standard deviation of that ofmean H4-6 chondrites (�17O � 0.73 � 0.09‰; n � 22;Clayton et al., 1991) and very similar to the mean of the two H3falls (Sharps and Dhajala) analyzed by Clayton et al.: 0.68� 0.04‰.

Because most material in lunar breccias is derived from localbedrock (McKay et al., 1991), it seems more likely that the fourknown C-rich chondritic clasts (all of which occur in H-chon-drite regolith breccias; Scott et al., 1981b) were derived fromthe H-chondrite asteroid than from an external source. In con-trast, CM2 chondrite clasts occur not only in H-chondriteregolith breccias (e.g., Plainview, Abbott, Tysnes Island, Ipi-ranga; Fodor and Keil, 1976; Fodor et al., 1976; Keil andFodor, 1980; Rubin, 1982), but also in CV3 Leoville (Keil etal., 1969), several howardites (e.g., Kapoeta, Bholghati, G’Day,Jodzie, LEW 85441, LEW 87015; Wilkening, 1973; Buchananet al., 1993; Zolensky et al., 1996) and the LEW 87295

Fig. 8. H-chondrite- and Mg-normalized abundance ratios of 25elements measured in clast PV1 by INAA. Lithophile elements areplotted at the left and siderophile and chalcophile elements at the right.In each group, elements are arranged from left to right in order ofdecreasing nebular condensation temperature. The mean refractorylithophile abundance ratio for 10 elements (Al–Lu) is 1.00�H-chon-drites. The semirefractory and common lithophiles are also unfraction-ated; moderately volatile Na and K are moderately depleted. Thesiderophile element pattern is more complex: refractory Ir is moder-ately enriched, the common siderophiles (Ni, Co, Fe) are depleted, andthe volatile siderophiles (Au, As, Ga) are essentially unfractionated.The two chalcophiles (Se, Zn) are moderately enriched.

polymict eucrite (Zolensky et al., 1996). The wide distributionCla

Hc a

dat


of CM2 clasts in regolith and near-surface fragmental brecciasfrom different asteroids indicates that they were derived froman external source.

4.1.2. Type 3 chondrite precursor

It is probable that PV1 started off as a type 3 chondritebecause of the presence of glass in some BO, PO and POPchondrules (e.g., Fig. 3a) and the juxtaposition of low-FeO andhigh-FeO chondrules (Fig. 1b). Although the overall olivineand low-Ca pyroxene compositional distributions (Fig. 2; fig. 4of Scott et al., 1981b) were affected by subsequent hydrother-mal alteration and partial metamorphic equilibration (see be-low), the composition of olivine cores in PV1 (which reflectnebular compositions more closely) has a minimum range ofFa1-19.

Normal primitive type 3 OC contain �10–15 vol% fine-grained, FeO-rich silicate matrix material (table 4 of Huss et al.,1981). In addition, �7% of their chondrules are RP and cryp-tocrystalline textural types (Nelson and Rubin, 2002). Althoughthe apparent absence of matrix material and dearth of RP andcryptocrystalline chondrules in PV1 is nominally inconsistentwith the hypothesis that this clast started off as a normal H3chondrite, altered forms of these materials appear to be present.We observed materials that we infer to be recrystallized clumpsof matrix material (i.e., I5; Fig. 5c), and thermally alteredfragments of RP and cryptocrystalline chondrules (i.e., (i.e.,Figs. 5a,b). This is consistent with the hypothesis that PV1silicates were normal type 3 OC material before impact heating(see below).

We also note that the C-isotopic composition of PV1(�13CPDB � �21‰; Scott et al., 1988) is marginally within therange of type 3 OC falls (�20.6 to �27.5‰; mean �13CPDB

� �23.6 � 2.0‰; n � 10; Grady et al., 1982) but outside therange of type 4–6 OC (typically �25 to �30‰; Grady et al.,1982). This is consistent with C in PV1 having the samegeneral provenance as C in type 3 OC.

4.2. Impact Heating

If PV1 was initially a normal H3 chondrite, it would havehad �18 wt% metallic Fe-Ni and �5.4 wt% troilite(Jarosewich, 1990). In contrast, modal analysis indicates thatPV1 contains ��6 vol% (i.e., ��14 wt%) metallic Fe-Ni, and�7 vol% (i.e., �10 wt%) troilite (Scott et al., 1981b); thesephases appear to be heterogeneously distributed (Fig. 1c). Themodal abundances of metal and troilite in the clast were prob-ably enhanced by the inclusion in the modal analysis of late-formed vein materials. These materials may have had a hightroilite/metal ratio, possibly similar to that of a eutectic mixture(7.5:1). The clast thus lost a significant fraction of the originalH3 metal; it must also have lost appreciable troilite. Moderatedepletion of metallic Fe-Ni in PV1 is indicated by the Mg- andH-chondrite-normalized abundance ratios of the common sid-erophile elements Fe (0.67), Ni (0.70) and Co (0.43) in the bulkanalysis (Table 1; Fig. 8). The possibility of significant loss ofmetal from PV1 due to terrestrial weathering can be discountedbecause the surrounding Plainview host is not depleted inmetallic Fe-Ni (Fig. 1).

Loss of metal requires a high-temperature event wherein

temperatures reached or exceeded the Fe-FeS eutectic (988°C;Kullerud, 1963) and where shear forces separated metal-sulfideliquid from silicate solids. This event is likely to have beenimpact heating on the parent body with associated separation ofa metal-sulfide melt. Many impact-melt-rock clasts in chon-dritic meteorites are highly depleted in metallic Fe-Ni andtroilite (e.g., Fodor et al., 1972; Fodor and Keil, 1976; Wilk-ening, 1978). The somewhat heterogeneous distribution ofmetal and sulfide in PV1 suggests that some portions of theclast may not have reached the Fe-FeS eutectic. Quenchingmay have preserved the glass in some of the chondrules (par-ticularly in those regions that may have experienced somewhatlower maximum temperatures).

The siderophile/chalcophile elemental abundance pattern inPV1 is complex (Fig. 8). There are moderate depletions ofcommon siderophiles, relatively unfractionated abundances ofvolatile siderophiles, a moderate enrichment in refractory Ir,and (consistent with the moderately high modal abundance oftroilite) moderate enrichment in the chalcophiles Se and Zn.This complex pattern may partly reflect sampling variations,loss of some metal and sulfide during the impact event, and,probably, late-stage introduction of metal and sulfide fromoutside the clast.

There is additional evidence for a high-temperature event.Our BSE images of PV1 reveal that most interfaces betweenadjacent chondrules are fused together (Fig. 4), rendering theclast somewhat analogous to a conglomerate of compound-chondrule clusters wherein each cluster consists of 20 indi-vidual chondrules. Such chondrule fusion requires temperaturesthat exceeded the solidus temperature for H chondrites, whichis �1120°C (Jurewicz et al., 1995) and probably closer to1100°C. Melting and partial loss of feldspathic components(including chondrule glass), possibly combined with loss of avolatile-rich vapor, is consistent with the depletion of moder-ately volatile lithophile elements in PV1: the Mg- and H-chondrite-normalized abundance ratios of Na and K are 0.75and 0.70, respectively (Table 1; Fig. 8).

PV1 also contains several objects (e.g., I5; Fig. 5c) thatappear to be recrystallized clumps of fine-grained silicate ma-trix material. We suggest that the apparent absence in PV1 offine-grained silicate-rich matrix material is a consequence ofthe impact heating and recrystallization of this material.

The recrystallized RP and cryptocrystalline chondrules (e.g.,objects E6, D2) in PV1 are also consistent with shock heatingof the clast.

By analogy with porous, melt-bearing fragmental brecciasfound in impact-crater ejecta deposits (Stöffler et al., 1980), itseems plausible that, at this stage, the PV1 precursor lithologyhad a high porosity. The voids may have been the sites whereC-rich material was later incorporated (see below).

It has been suggested that PV1 could be a heterogeneousmixture of materials with different heating histories. However,the occurrence of clusters of fused chondrules throughout theclast attests to fairly uniform heating.

In an attempt to circumvent this difficulty, Brearley sug-gested that the melded chondrule boundaries were not causedby fusion at high temperatures, but rather may have resultedfrom pressure solution as is the case for ooids (a.k.a. ooliths—millimeter-size, spheroidal inorganic precipitates) in some ter-
restrial sedimentary rocks (fig. 2 of Skinner, 1989). However,


some of the ooids shown by Skinner appear to be fragmented,a likely occurrence in the shallow, wave-agitated water whereooids are produced. Some of the apparently conjoined ooidscould be juxtaposed fragments that were jostled into placerather than melded together by pressure solution. In any case,low-temperature pressure solution is less likely to affect maficminerals in chondrules than calcium carbonate in ooids. Fur-thermore, the melded chondrules in PV1 differ significantlyfrom melded ooids in being recrystallized in regions away fromboundary surfaces.

4.3. Hydrothermal Alteration

The typical small euhedral olivine grains within compoundchondrule G7 in PV1 consist of 3–5-�m-size low-FeO coresrimmed by ferroan olivine (Fig. 6d). Narrow fractures in manyof these grains are also filled with ferroan olivine. The grainsclosely resemble the altered olivine grains in amoeboid olivineinclusions (AOIs) in CO3.2–3.5 chondrites wherein ferroanolivine conforms to the shapes of preexisting cracks and to theoutlines of grain boundaries (Chizmadia et al., 2002). Numer-ous coarser olivine grains within porphyritic chondrulesthroughout PV1 (e.g., Fig. 6) are analogous to these smallgrains: they also consist of low-FeO cores surrounded by thickpatches of ferroan olivine.

The ferroan olivine patches in PV1 were probably caused byhydrothermal alteration at low water/rock ratios. The ferroanolivine rims are more fayalitic than the average for equilibratedH chondrites (Fa17.3–20.2; Rubin, 1990); thus, some form ofalteration process must be responsible for their formation. Thisprocess is not likely to be aqueous alteration sensu strictobecause of the elevated temperatures, the apparent absence ofhydrous phases in the clast, and the preservation of chondruleglass. It seems probable that some metallic Fe was oxidized andthat the resultant FeO, combined with SiO2 derived from fine-grained phases, was transported by a fluid along grain bound-aries and through preexisting fractures in the low-FeO olivinegrains; it was at these sites that fayalitic olivine precipitated.

Temperatures were high enough in PV1 to allow minorvolume diffusion in the olivine: Fe�2 from the newly precipi-tated fayalitic olivine exchanged partially with Mg�2 fromadjacent low-FeO olivine to form diffusive haloes of ferroanolivine around the low-FeO olivine cores. The effective dis-tance for volume diffusion in PV1 olivine during this event is�1–2 �m as indicated by the thickness of the ferroan olivinerinds around low-FeO cores in the rim of BO chondrule B4(Fig. 3a).

There is a paucity of olivines of composition Fa0-3 in H/L3.6Tieschitz (fig. 1 of Dodd et al., 1967) presumably caused bypartial equilibration of olivine during metamorphism. Simi-larly, the olivine compositional distribution in PV1 shows apaucity of Fa0-2 olivine grains (Fig. 2a), presumably also dueto partial equilibration. In contrast, histograms of olivine dis-tributions in more primitive type 3 OC (e.g., LL3.0 Semarkona)are bimodal with a sharp peak at low-FeO compositions (e.g.,fig. 1 of Takagi et al., 2004).

The heat responsible for hydrothermal alteration and diffu-sion may be residual shock heat that resulted from impact
heating and burial of the clast.
4.4. Fracturing

The numerous 20–160-�m-long fractures that transect everychondrule in PV1 (e.g., Fig. 3b) were caused by crushing,presumably also associated with an impact event. Most frac-turing postdated impact heating; if fracturing had precededimpact heating, it seems likely that the fractures in recrystal-lized silicate matrix object I5 in PV1 would have been largelyobliterated.

4.5. Potential Sources of Carbon

We suggest two potential source regions for the C-richmaterial in the chondritic clasts: the solar nebula and theH-chondrite parent asteroid. We have discussed the evidenceindicating that the PV1 precursor was initially a normal H3chondrite. Because type 3 OC contain ��1 wt% C(Jarosewich, 1990), we think it unlikely that the C-rich clastsobtained their abundant C during nebular agglomeration. Wetherefore conclude that the clasts acquired their C on theH-chondrite asteroid.

Two mechanisms that could potentially account for C en-richment on the parent body are (1) outgassing of CO from theasteroid interior and (2) introduction of C from a C-rich pro-jectile.

4.5.1. Outgassing

It is conceivable that the C was derived by outgassing of COfrom metamorphosed (H4-6) rocks in the asteroid interior fol-lowed by dissociation of the CO to form C within permeablematerials near the asteroid surface (Sugiura et al., 1986). Onepossible reaction is:

Fe(s) � CO(g) � MgSiO3(s) →FeMgSiO4(s) � C(s) (1)

In this reaction, oxidation of 10 wt% Fe would yield �2 wt%C. This value is higher than that in type 3 OC (which typicallycontain ��1 wt% C), but much lower than that in PV1 (13wt%). Although PV1 is depleted in metallic Fe relative tonormal H-group chondrites by �4 wt%, a portion of thisdepletion is attributable to loss of a metal-rich liquid during animpact event. Thus, the moderately low concentration of Fe inPV1 cannot be attributed solely to oxidation of Fe as shown inEqn. 1. In addition, thermodynamic data indicate that at T�1100 K and pCO �1 atm, the reverse reaction would spon-taneously occur.

4.5.2. C-rich projectile

Another possibility is that the C was delivered as carbona-ceous matter derived from a C-rich projectile. The most C-richplanetesimal-size projectiles are comets. Jessberger et al.(1988) reported that the dust � ice fraction of Comet Halleycontains �19 wt% C, although some estimates of the C con-tents of comets are appreciably lower (e.g., �8 wt%;Delsemme, 1982). Carbon occurs in comets in a wide variety ofphases including simple, volatile species (e.g., CO, CO2), com-plex, multiatomic species (e.g., HCOOCH3, HOCH2CH2OH—ethylene glycol), organic (CHON) grains, and organic refrac-
tory coatings on silicate grains (e.g., Mumma et al., 1993; Rahe


et al., 1995; Greenberg, 1998). By analogy with CI and CMchondrites, complex hydrocarbons and kerogen (Cronin et al.,1988) that cannot be detected by remote studies are probablypresent as well. The high-molecular-weight compounds tend tobe relatively refractory and would likely be the most importantprecursors of the carbon in PV1.

It seems possible that the hypothetical cometary projectileserved as the source of the fluid responsible for the hydrother-mal alteration of PV1, but the silicate portion of the comet isnot a recognizable component of the clast.

It is not unreasonable that a cometary impact could haveaffected the chondritic clasts. A small fraction of Jupiter-familycomets (JFCs) evolve into near-Earth objects (Levison andDuncan, 1997) and could impact main-belt asteroids at rela-tively low velocities of �5 km s�1 (similar to those of averageasteroid/asteroid collisions) (Campins and Swindle, 1998). Re-cent studies estimate that �1% of the projectiles impactingasteroids today are JFCs (Wetherill, 1978; Near-Earth ObjectScience Definition Team, 2003).

It is conceivable that the carbon in the C-rich chondriticclasts was derived from liquid hydrocarbons that flowed intovoids in the clast. Such liquids could have been subsequentlybroken down into carbonaceous matter with high C/H ratios bydehydrogenation processes perhaps resulting from heat gener-ated by the same impact event(s) that fractured and faulted theclast. The �13CPDB composition of PV1 (�21‰; Scott et al.,1988) is within the wide range reported for comets (�11‰ to�43‰; Wyckoff et al., 2000) as well as that of type 3 OC falls(�20.6 to �27.5‰; Grady et al., 1982).

4.6. Inferred History of PV1

A volume of H3 material, residing in the near-surface regionof the H-chondrite asteroid, was impact heated. The nature ofthe projectile responsible for this event is unknown. Tempera-tures in the clast reached �1100°C and chondrules throughoutPV1 fused together into clusters. Fine-grained ferroan silicatematrix material was recrystallized, and, in some regions of theclast, fine-grained RP and cryptocrystalline chondrule texturaltypes were similarly affected. Some metal and sulfide meltedand separated from solid silicate, and the rock was left withnumerous impact-induced voids.

At a later time, a C-rich projectile accreted to the parentasteroid; the PV1 precursor was fractured during this event.Carbon from the (presumably cometary) projectile filled thevoids; the carbon may have been transported as a fluid that laterbroke down into elemental C. The impact of this projectileprobably caused moderate heating of the clast. Annealing at300 to 450°C caused the graphitizable organic C in PV1 topyrolyze and dehydrogenate, thereby transforming it intopoorly graphitized C (Brearley, 1990). Additional impactevents caused fracturing in the clast and surrounding materialsin the Plainview regolith.

While the clast remained at elevated temperatures, hydrother-mal alteration at low water/rock ratios caused partial equilibrationof PV1, resulting in the paucity of forsterite grains in the olivinecompositional distribution. This process also caused oxidation ofsome metallic Fe; FeO combined with SiO derived from fine-
2
grained phases and was transported by a fluid through fractures

and along grain boundaries where it precipitated as fayalite. Fer-roan olivine haloes formed around the low-FeO olivine cores.

Before PV1 was incorporated into the Plainview host, thehost was thermally metamorphosed to petrologic type 5, reach-ing temperatures of �500 to 600°C. If the Plainview host hadbeen metamorphosed after the C-rich PV1 clast was incorpo-rated, chondrule glass in PV1 would have recrystallized.

A subsequent impact event (or series of events) dislodged thePV1 clast from its formation region and deposited it into its finallocation in the H-chondrite regolith. Other C-rich chondritic clastswere deposited elsewhere in the regolith. This is indicated by thedifferent cosmic-ray exposure ages of the regolith breccias thatcontain these clasts (Plainview, 6.0 Ma; Dimmitt, 6.0 Ma; Weston,35.5 Ma; Graf and Marti, 1995), demonstrating that the brecciaswere derived from different locations on the parent asteroid. Afterbeing incorporated into its final regolithic location, one or moreadditional impact event(s) caused extensive faulting in the Plain-view host and PV1 clast.

Some of the C-rich aggregates that occur in the relatively C-richtype 3 OC (e.g., Sharps, ALHA77011) may have been derivedfrom disrupted C-rich chondritic clasts similar to PV1 (Scott et al.,1981a). However, the abundant metallic Fe-Ni and minor chro-mite in most C-rich aggregates (Scott et al., 1988; Brearley, 1990)indicate different sources or more complex histories.

Acknowledgments—We thank the Institute of Meteoritics at the Uni-versity of New Mexico for the chip of clast PV1 and the loan ofPlainview thin section UNM 273. We gratefully acknowledge infor-mative discussions with E. R. D. Scott and A. J. Brearley. The articlebenefited from thorough reviews by E. R. D. Scott, M. E. Zolensky, andparticularly A. J. Brearley. This work was supported in part by NASAgrants NAG5-12967 (A. E. Rubin) and NAG5-12887 (J. T. Wasson).J. M. Trigo-Rodriguez thanks the Spanish Secretary of Education andUniversities for a postdoctoral grant.

Associate editor: M. Grady

REFERENCES

Berkley J. L., Brown H. G., Keil K., Carter N. L., Mercier J-C. C., andHuss G. (1976) The Kenna ureilite: An ultramafic rock with evi-dence for igneous, metamorphic and shock origin. Geochim. Cos-mochim. Acta 40, 1429–1437.

Bischoff A., Rubin A. E., Keil K., and Stöffler D. (1983) Lithificationof gas-rich chondrite regolith breccias by grain boundary andlocalized shock melting. Earth Planet. Sci. Lett. 66, 1–10.

Brearley A. J. (1990) Carbon-rich aggregates in type 3 ordinary chon-drites: Characterization, origins and thermal history. Geochim. Cos-mochim. Acta 54, 831–850.

Brearley A. J., Scott E. R. D., and Keil K. (1987) Carbon-rich aggre-gates in ordinary chondrites: Transmission electron microscopyobservation of Sharps (H3) and Plainview (H regolith breccia)(abstract). Meteoritics 22, 33819.

Buchanan P. C., Zolensky M. E., and Reid A. M. (1993) Carbonaceouschondrite clasts in the howardites Bholghati and EET 87513. Me-teoritics 28, 659–669.

Campins H. and Swindle T. D. (1998) Expected characteristics ofcometary meteorites. Meteorit. Planet. Sci. 33, 1201–1211.

Chizmadia L. J., Rubin A. E., and Wasson J. T. (2002) Mineralogy andpetrology of amoeboid olivine inclusions in CO3 chondrites: Re-lationship to parent-body aqueous alteration. Meteorit. Planet. Sci.37, 1781–1796.

Clayton R. N., Mayeda T. K., Goswami J. N., and Olsen E. J. (1991)Oxygen isotope studies of ordinary chondrites. Geochim. Cosmo-chim. Acta 55, 2317–2337.

Cronin J. R., Pizzarello S. and Creikshank D. P. (1988) Organic matterin carbonaceous chondrites, planetary satellites, asteroids and com-


ets. In Meteorites and the Early Solar System (eds. J. F. andKerridge M. S. Matthews), pp. 819–857. University of ArizonaPress, Tucson.

Delsemme A. H. (1982) Chemical composition of cometary nuclei. InComets (eds. L. L. and Wilkening M. S. Matthews), pp. 85–130.University of Arizona Press, Tucson.

Dodd R. T., Van Schmus W. R., and Koffman D. M. (1967) A surveyof the unequilibrated ordinary chondrites. Geochim. Cosmochim.Acta 31, 921–951.

Fodor R. V., Keil K., and Jarosewich E. (1972) The Oro Grande, NewMexico, chondrite and its lithic inclusion. Meteoritics 7, 495–507.

Fodor R. V. and Keil K. (1976) Carbonaceous and noncarbonaceouslithic fragments in the Plainview, Texas chondrite: Origin andhistory. Geochim. Cosmochim. Acta 40, 177–189.

Fodor R. V., Keil K., Wilkening L. L., Bogard D. D., and Gibson E. K.(1976) Origin and history of a meteorite parent-body regolithbreccia: Carbonaceous lithic fragments in the Abbott, New Mexico,chondrite. Special Pub. New Mexico Geol. Soc. 6, 206–218.

Graf T. and Marti K. (1995) Collisional history of H chondrites. J.Geophys. Res. 100, 21247–21263.

Grady M. M. (2000) Catalogue of Meteorites. Cambridge UniversityPress.

Grady M. M., Swart P. K., and Pillinger C. T. (1982) The variablecarbon isotopic composition of type 3 ordinary chondrites. Proc.Lunar Planet. Sci. Conf. 13, A289–A296.

Greenberg J. M. (1998) Making a comet nucleus. Astron. Astrophys.330, 375–380.

Huss G. R., Keil K., and Taylor G. J. (1981) The matrices of unequili-brated ordinary chondrites: Implications for the origin and historyof chondrites. Geochim. Cosmochim. Acta 45, 33–51.

Jarosewich E. (1990) Chemical analyses of meteorites: A compilationof stony and iron meteorite analyses. Meteoritics 25, 323–337.

Jessberger E. K., Christoforidis A., and Kissel J. (1988) Aspects of themajor element composition of Halley’s dust. Nature 332, 691–695.

Jones R. H. (1996) Relict grains in chondrules: Evidence for chondrulerecycling. In Chondrules and the Protoplanetary Disk (eds. R. H.Hewins, R. H., and Jones E. R. D. Scott), pp. 163–172. CambridgeUniversity Press. USA.

Jurewicz A. J. G., Mittlefehldt D. W., and Jones J. H. (1995) Experi-mental partial melting of the St. Severin (LL) and Lost City (H)chondrites. Geochim. Cosmochim. Acta 59, 391–408.

Kallemeyn G. W., Rubin A. E., Wang D., and Wasson J. T. (1989)Ordinary chondrites: Bulk compositions, classification, lithophile-element fractionations and composition-petrographic type relation-ships. Geochim. Cosmochim. Acta 53, 2747–2767.

Keil K., Huss G. I. and Wiik H. B. (1969) The Leoville, Kansas,meteorite: A polymict breccia of carbonaceous chondrites andachondrite (abstract). In Meteorite Research (ed. P. M. Millman), p.217. D. Reidel, Dordrecht.

Keil K. and Fodor R. V. (1980) Origin and history of the polymict-brecciated Tysnes Island chondrite and its carbonaceous and non-carbonaceous lithic fragments. Chem. Erde 39, 1–26.

Keil K., Fodor R. V., Starzyk P. M., Schmitt R. A., Bogard D. D., andHusain L. (1980) A 3.6-b.y.-old impact-melt rock fragment in thePlainview chondrite: Implications for the age of the H-group chon-drite parent body regolith formation. Earth Planet. Sci. Lett. 51,235–247.

Kullerud G. (1963) The Fe-Ni-S system. Carnegie Inst. Wash. Yr. Bk.62, 175–189.

Levison H. F. and Duncan M. J. (1997) From the Kuiper Belt toJupiter-family comets: The spatial distributions of ecliptic comets.Icarus 127, 13–32.

McKay D. S., Heiken G., Basu A., Blanford G., Simon S., Reedy R.,French B. M. and Papike J. (1991) The lunar regolith. In: LunarSourcebook: A User’s Guide to the Moon (eds. G. H. Heiken, D. T.and Vaniman B. M. French), pp. 285–356. Cambridge UniversityPress. USA.

McKinley S., Scott E. R. D., Taylor G. J., and Keil K. (1981) A uniquetype 3 chondrite containing graphite-magnetite aggregates—AllanHills A77011. Proc. Lunar Planet. Sci. 12B, 1039–1048.

McSween H. Y. and Richardson S. M. (1977) The composition ofcarbonaceous chondrite matrix. Geochim. Cosmochim. Acta 41,
1145–1161.
Mostefaoui S., Perron C., Zinner E., and Sagon G. (2000) Metal-associated carbon in primitive chondrites: Structure, isotopic com-position and origin. Geochim. Cosmochim. Acta 64, 1945–1964.

Mumma M. J., Stern S. A. and Weisman P. R. (1993) Comets and theorigin of the solar system. In Protostars and Planets III (eds. E. H.and Levy J. L. Lunine), pp. 1177–1252. University of ArizonaPress, Tucson.

Near-Earth Object Science Definition Team (2003) Study to Determinethe Feasibility of Extending the Search for Near-Earth Objects toSmaller Limiting Diameters. NASA.

Nelson V. E. and Rubin A. E. (2002) Size-frequency distributions ofchondrules and chondrule fragments in LL3 chondrites: Implica-tions for parent-body fragmentation of chondrules. Meteorit.Planet. Sci. 37, 1361–1376.

Rahe J., Vanysek V. and Weissman P. R. (1995) Properties of cometarynuclei. In Hazards Due to Comets and Asteroids (ed. T. Gehrels),pp. 597–634. University of Arizona Press, Tucson.

Rubin A. E. (1982) Petrology and origin of brecciated chondriticmeteorites. Ph.D. thesis. University of New Mexico, Albuquerque.

Rubin A. E. (1990) Kamacite and olivine in ordinary chondrites:Intergroup and intragroup relationships. Geochim. Cosmochim.Acta 54, 1217–1232.

Rubin A. E. (2000) Petrologic, geochemical and experimental con-straints on models of chondrule formation. Earth Sci. Rev. 50, 3–27.

Schultz L. and Kruse H. (1989) Helium, neon and argon in meteor-ites—A data compilation. Meteoritics 24, 155–172.

Scott E. R. D., Taylor G. J., Rubin A. E., Okada A., and Keil K. (1981a)Graphite-magnetite aggregates in ordinary chondritic meteorites.Nature 291, 544–546.

Scott E. R. D., Rubin A. E., Taylor G. J., and Keil K. (1981b) New kindof type 3 chondrite with a graphite-magnetite matrix. Earth Planet.Sci. Lett. 56, 19–31.

Scott E. R. D., Brearley A. J., Keil K., Grady M. M., Pillinger C. T.,Clayton R. N., Mayeda T. K., Wieler R., and Signer P. (1988)Nature and origin of C-rich ordinary chondrites and chondriticclasts. Proc. Lunar Planet. Sci. Conf. 18, 513–523.

Skinner W. R. (1989) Compaction and lithification of chondrites (ab-stract). Lunar Planet. Sci. 20, 1020–1021.

Stöffler D., Knöll H.-D., Marvin U. B., Simonds C. H. and WarrenP. H. (1980) Recommended classification and nomenclature oflunar highland rocks—A committee report. In Proc. Conf. LunarHighlands Crust (ed. R. B. and Merrill J. J. Papike), pp. 51–70.Pergamon. USA.

Stöffler D., Keil K. and Scott E. R. D. (1991) Shock metamorphism ofordinary chondrites. Geochim. Cosmochim. Acta 55, 3845–3867.

Sugiura N., Arkani-Hamed J., and Strangeway D. W. (1986) Possibletransport of carbon in meteorite parent bodies. Earth Planet. Sci.Lett. 78, 148–156.

Takagi M., Huber H., Rubin A. E., and Wasson J. T. (2004) Distribu-tion of FeO/(FeO � MgO) in Semarkona chondrules: Implicationsfor chondrule formation and nebular evolution (abstract). Meteorit.Planet. Sci. 39:A103.

Wasson J. T. and Kallemeyn G. W. (1988) Compositions of chondrites.Phil. Trans. R. Soc. Lond. A 325, 535–544.

Wasson J. T., Krot A. N., Lee M. S., and Rubin A. E. (1995) Com-pound chondrules. Geochim. Cosmochim. Acta 59, 1847–1869.

Wetherill G. W. (1978) Dynamical evidence regarding the relationshipbetween asteroids and meteorites. In Asteroids: An ExplorationAssessment,pp. 17–35. Conf. Pub. 2053. NASA.

Wilkening L. L. (1973) Foreign inclusions in stony meteorites—I.Carbonaceous xenoliths in the Kapoeta howardite. Geochim. Cos-mochim. Acta 37, 1985–1989.

Wilkening L. L. (1978) Tysnes Island: An unusual clast composed ofsolidified, immiscible, Fe-FeS and silicate melt. Meteoritics 13,1–9.

Wyckoff S., Kleine M., Peterson B. A., Wehinger P. A., and ZiurysL. M. (2000) Carbon isotope abundances in comets. Astrophys. J.535, 991–999.

Zolensky M. E., Weisberg M. K., Buchanan P. C., and MittlefehldtD. W. (1996) Mineralogy of carbonaceous chondrite clasts in HED
achondrites and the moon. Meteorit. Planet. Sci. 31, 518–537.

carbon-rich chondritic clast pv1 from the plainview ... - uji · carbon-rich chondritic clast pv1...

Documents